Homologous recombination is an important genome-stabilizing force in mitotically dividing eukaryotic cells, where it serves an important DNA repair function and helps to insure the integrity of the genome prior to cellular division. Recombination does not involve naked DNA, but rather involves DNA that is highly packaged with proteins into chromatin. Not only is recombination potentially impacted by chromatin structure, but it also can be influenced by other DNA metabolic processes such as transcription and DNA replication. Mitotic recombination typically involves identical sequences on sister chromatids or on homologous chromosomes, but the presence of large amounts of repetitive DNA in eukaryotic genomes affords the opportunity for other types of interactions to occur. Interactions between dispersed repeated sequences can result in deleterious genome rearrangements such as inversions and translocations, and such rearrangements are causative factors in a number of human diseases. Cells thus have evolved a surveillance mechanism that enforces strict sequence identity requirements during recombination, and this mechanism involves anti-recombination activity of the mismatch repair machinery. The studies outlined in this proposal will utilized the yeast Saccharomyces cerevisiae to examine two aspects of mitotic recombination: (1) how the mismatch repair machinery prevents recombination between non-identical sequences and (2) how high levels of transcription stimulate recombination. Interactions between non-identical sequences will be studied using artificial recombination substrates, and a variety of genetic methods will be employed to elucidate the mechanism of mismatch repair-associated anti-recombination. In addition, recombination systems will be developed that will allow the impact of sequence divergence on the resolution of recombination events to be examined, as the resolution step determines whether or not the event will impact genome structure. With regard to the transcription-stimulated recombination, it has been speculated that high levels of transcription either increase the frequency of recombination-initiating DNA lesions, or facilitate the interactions of recombination proteins with highly packaged DNA. These two hypotheses will be tested using genetic approaches.